[Ndca-l] Issues concerning the high school community

Stefan Bauschard stefan.bauschard
Wed Dec 18 19:35:01 EST 2013


Post by Gary Larson on MPJ.  I post it here because he already posted it
publicly on the CEDA forums and because I strongly agree with it -- it is
one of the reasons that I support a large number of categories (I really
support ordinals but those have not caught on in HS debate).

Started by gabemurillo<http://www.cedadebate.org/forum/index.php?action=profile;u=221>
-
Last post by glarson<http://www.cedadebate.org/forum/index.php?action=profile;u=18>
As the acknowledged perpetrator behind the current MPJ system, let me break
my silence by offering one warning regarding potential solutions.

Almost everyone who is critiquing the current model of MPJ concurs that it
is a problem to attempt to maximize preference (whether or not it is more
or less important than maximizing mutuality).  While the reasons might
differ, many different voices are advocating that we increase the range of
judges that any given team is judged by AND increase the range of teams
that any give judge is assigned to judge.

Assuming that we agree that these are appropriate objectives, there are two
proposed ways to accomplish it.  Most of the proposed alternatives choose
to increase randomness, either by reducing the number of categories used,
increasing the size of each category and/or by reducing the impact of
preference to something closer to zero (Wake's original experiment).  The
presumption is that increased randomness results in increased variability
which in turn results in increased diversity (race, gender, ethnicity,
argument preferences, etc.).

On one level, this approach accomplishes some of its goals.  It is the case
overall that a team is more likely to receive a wider range of judges and a
judge is more likely to receive a wider range of teams the broader the
categories and the more random the system becomes.  BUT, we need to be
aware that while the overall results are more diverse, the variability of
results between any two teams is even more variable because of the small
sample size.  To explain, once upon a time we had random pairings for
teams.  But we observed that what counted as random overall still meant
that one team could face all of their opponents from the top 10% of the
field while another team could face all of their opponents from the bottom
10%.  Even in the world where we might have only two random presets,
experience shows that without pre-seeding or assigning ratings, some team
could "randomly" get paired against two teams vying for a Copeland.

The same thing happens the more "random" or "broadly categorical" judge
assignment becomes.  Yes it is true that we will have more debates in which
a team gets a judge that they would have ranked in the 50's, 60's, 70's or
80's, but that won't happen at all uniformly.  We will randomly get some
teams who get 8 judges they prefer highly or 8 rounds in which they prefer
the judge significantly more than their opponent while we will get some
other team that has exactly the opposite outcome.  That's an inevitable
outcome of an experiment like Wake's original model or Strauss's return to
a three category system (circle, neutral, strike or ABX).  So while it will
be meet some objectives for some teams and some judges, for other teams
and/or judges it won't and more critically, it will reduce the "fairness"
across the range of teams in a tournament.  Random acts happen very
capriciously.

To be honest, I worry about the random (and non-random) variability that
already happens using current MPJ models.  We worry about outcomes where
the "luckiest" team averages 8.5 and the "unluckiest" team average 18.5.
We worry about whether some teams more frequently get judges they prefer
more highly and whether it has an impact on the outcome.  Algorithms work
to balance team pref across rounds and keep mutuality close enough to not
make a difference.  But as soon as we permit randomness to take a bigger
role, either by reducing the role of preference or by increasing the size
of categories, the difference between the lucky and the unlikely will grow
simultaneously.  In my judgment, that's a problem.

So does that mean we're stuck with the status quo?  Not at all.  Rather
than trying to increase variability/diversity by increasing randomness, I
would favor making diversity an explicit objective of a decidedly
non-random procedure.  The algorithms can be used to create "mutual
non-preferred" judging in selected rounds, where the system systematically
and somewhat uniformly attempts to increase diversity for all participants
by intentionally reversing the polarity of preference in some rounds (up to
some agreed upon threshold).

Before leaving the topic, let me make a couple of other observations about
randomness, categories, ordinal prefs, etc.  I strongly suspect that
completing a pref sheet has pyschological impacts that may prove to be even
more important than the empirical results.  I believe that Gordon (and
others) suspect that filling out an ordinal sheet has a different
psychological impact than filling out a categorical sheet and that that
impact outweighs other factors.  He might be right.  We may assume an
inappropriate sense of "power" in ranking each judge ordinally as opposed
to clustering them into groups.  We may assume more "precision" than is
warranted by the pairing algorithm.  I think that those factors need to be
evaluated independent of whether one system or another better maximizes our
judge assignment objectives.

I also suspect that the more "precision" (real or otherwise) invested in an
ordinal sheet raises significant questions about agency.  While I can
identify the real lack of equal outcomes in random events, we often prefer
to be controlled by "karma."  One of the problems teams face at present is
that they fear that the current system isn't fair because other teams do a
better job of controlling the outcome ("gaming") than they do.  Some
suspect that it is a systematic advantage by one type of team or another.
Others fear that somebody else knows something about the "deep" logic of
the system that they don't know that gives them an advantage.  Even if I
could prove that nobody has a systematic advantage or a better strategy of
maximizing their outcomes, it doesn't eliminate the angst.  Randomness
levels that playing field.  So even if my disadvantage proves to be even
greater due to the cruel hand of fate, I can take bad luck better than I
can deal with a fear that I'm getting played.


On Wed, Dec 18, 2013 at 4:26 PM, Robb Gray <firstpacific at att.net> wrote:

> This is unfortunate. If we all had an email list and the ability to only
> send it directly to the people on that list this would be less likely. I
> propose that the NDCA set up a list where all communications by design are
> supposed to stay purely among the people on the list. If anyone wants to
> opt
> out they can but if you communicate to the people on the list you should
> have a reasonable expectation that your communication stays only with the
> people on the list. Certainly if I sent a private, or 100 private emails
> (all the same) I could have a reasonable expectation that it would stay
> that
> way. Best Regards, Robb Gray
>
> -----Original Message-----
> From: ndca-l-bounces at lists.debatecoaches.org
> [mailto:ndca-l-bounces at lists.debatecoaches.org] On Behalf Of Chris Palmer
> Sent: Wednesday, December 18, 2013 9:47 AM
> To: Jim Menick
> Cc: ndca-l at lists.debatecoaches.org
> Subject: Re: [Ndca-l] Issues concerning the high school community
>
>
> Yes, the emails posted to this list are archived publicly here:
>
> http://lists.debatecoaches.org/pipermail/ndca-l-debatecoaches.org/
>
> So please do not post anything to this list you wouldn't want on the web
> anyway; it's going to land there one way or another.
>
> Cheers,
> -Chris
>
> ---
> Free Tournament Management software from IDEA:  http://www.tabroom.com
> Asst Coach, Lexington Debate: http://www.lexdebate.org
> LD & Congress tournament director, NDCA:  http://www.debatecoaches.org
>
>
> On 18/12/13, Jim Menick wrote:
> > I've been porting it over the the website. I'm pretty sure this list is
> > open to anyone. I don't know about how or when folks will speak their
> > minds, but I do find emails have a tendency to disappear. Plus I'm seeing
> > listserv responses in a number of different gmail windows rather
> > arbitrarily (which is, as gmail people know, the bane of the latest
> > iteration), making it hard to follow.
> >
> > jimmenick.com: Debate, books, and various nonsense
> >
> >
> > On Wed, Dec 18, 2013 at 11:06 AM, Stefan Bauschard <
> > stefan.bauschard at gmail.com> wrote:
> >
> > > I like the fact that this conversation is occurring on email because
> this
> > > listserv is, I believe, limited to debate coaches. I think people are
> more
> > > likely to speak what is on their minds than if this discussion is moved
> to
> > > a public message board.
> > >
> > > Also, I'm a bit sensitive to coaches airing all of their disagreements
> > > (which is important to do) in front of all of the students.
> > >
> > > I'm curious as to what others think.
> > >
> > >
> > > On Wed, Dec 18, 2013 at 9:48 AM, John M. Masslon II
> <johnmasslon at gmail.com
> > > > wrote:
> > >
> > >>
> > >>
> > >>
> > >>
> > >>
> > >>
> > >> *I think this is by far the most pressing issue facing HS debate at
> the
> > >> moment and am glad to see that a discussion has begun regarding our
> > >> community's response.  I wish I could be as eloquent and thorough as
> > >> Manuel, Stefan, Aaron, Tara, et al, but I know that will not happen.
>  I
> > >> would, however, like to add a few of my thoughts on the subject.
> First,
> > >> I think we need to begin is by looking in the mirror.  I believe that
> many
> > >> of the problems that have occurred in college debate are the result of
> > >> coaches pushing the unethical, and in some cases illegal, behavior by
> their
> > >> students.  We only need to look as far as the CEDA quarterfinal round
> a
> few
> > >> years ago to see that we, as coaches, are the ones that act in a
> manner
> > >> that makes debate a hostile environment for fellow coaches and our
> > >> students.  This is unacceptable.  I believe that the first step we
> should
> > >> take in ensuring that there is no bullying, harassment, etc. by our
> > >> students is to take forceful action against coaches that promote such
> > >> behavior.   I know this will not be a popular sentiment, but I think
> we
> > >> need to ban coaches who promote such behavior.  I run the tab room at
> the
> > >> biggest HS tournament in Western Pennsylvania, and I would never allow
> > >> Shanara Reid-Brinkley to step foot in a debate at my tournament.  If
> she
> > >> chose to coach a HS team, I would not allow that team to be entered in
> the
> > >> tournament.  The same would hold true for Bill Shanahan.  By
> eliminating
> > >> coaches that act in such a manner from our activity, we ensure that
> these
> > >> our students are not being influenced by such bad examples.  We set a
> > >> precedent for our students that if you act in a manner that is
> inconsistent
> > >> with the behavioral norms of debate, you are gone.  Thus, let us begin
> by
> > >> looking in the mirror. I also think that it is important that we
> ensure
> > >> that students who violate the norms of acceptable behavior at debate
> > >> tournaments are punished accordingly.  For example, if a student tells
> > >> his/her opponent in a round that they should kill themselves . that
> student
> > >> should be done with HS debate.  That student's coach should be the one
> to
> > >> enforce such a rule, however, that is probably unrealistic.  Instead,
> I
> > >> think that if a student behaves in such a manner at a tournament, the
> > >> tournament director should inform all other tournament directors to
> not
> > >> accept that student into future tournaments.  That way, a student
> knows
> > >> that if their actions fall so far outside the bounds of acceptable
> > >> behavior, they will no longer be permitted to participate in this
> great
> > >> activity.   Finally, I think that the tabula rasa paradigm has gone to
> far.
> > >>  I have traditionally been a fan of "anything goes" debate and have
> voted
> > >> for "performance" teams more than I have voted against such teams.
> > >>  However, after what we have seen in college debate recently, I have
> > >> decided that enough is enough.  I have decided that I will no longer
> vote
> > >> for such a team and will give them 0 speaks.  I hope that this stops
> the
> > >> spread of what has killed college debate into the HS realm.  I'm not
> naive
> > >> enough to think that my one person stand will have such an impact, but
> I
> > >> hope that after the inevitable collapse of college debate, more people
> will
> > >> realize we need to return to a more traditional style of debate.  This
> does
> > >> not mean that we need to get rid of critical arguments, it just means
> that
> > >> such arguments need to be made in the context of the resolution.   *
> The
> > >> views expressed above are my own and do not reflect the views of
> Bethel
> > >> Parks SD, Bethel Park SHS, or my employer.   John M. Masslon II, Esq.
> > >> Policy Debate Coach, Bethel Park SHS *
> > >>
> > >> phone: 703-791-9483
> > >> fax: 866-241-5219
> > >> e-mail: johnmasslon at gmail.com
> > >>
> > >>
> > >>
> > >> _______________________________________________
> > >> Ndca-l mailing list
> > >> Ndca-l at lists.debatecoaches.org
> > >> http://lists.debatecoaches.org/listinfo.cgi/ndca-l-debatecoaches.org
> > >>
> > >>
> > >
> > >
> > > --
> > > Stefan Bauschard
> > >
> > > Debate Coach, Harvard Debate
> > > Debate Coach, Lakeland Schools
> > > International Programs Consultant, National Forensic League (US)
> > > Consultant, National Forensic League of Korea (ROK)
> > > Consultant, Dipont Education Management (China)
> > >
> > > (C) 781-775-0433
> > > (F) 617-588-0283
> > >
> > >
> > >
> > > _______________________________________________
> > > Ndca-l mailing list
> > > Ndca-l at lists.debatecoaches.org
> > > http://lists.debatecoaches.org/listinfo.cgi/ndca-l-debatecoaches.org
> > >
> > >
>
> > _______________________________________________
> > Ndca-l mailing list
> > Ndca-l at lists.debatecoaches.org
> > http://lists.debatecoaches.org/listinfo.cgi/ndca-l-debatecoaches.org
>
>
> _______________________________________________
> Ndca-l mailing list
> Ndca-l at lists.debatecoaches.org
> http://lists.debatecoaches.org/listinfo.cgi/ndca-l-debatecoaches.org
>
> _______________________________________________
> Ndca-l mailing list
> Ndca-l at lists.debatecoaches.org
> http://lists.debatecoaches.org/listinfo.cgi/ndca-l-debatecoaches.org
>



-- 
Stefan Bauschard

Debate Coach, Harvard Debate
Debate Coach, Lakeland Schools
International Programs Consultant, National Forensic League (US)
Consultant, National Forensic League of Korea (ROK)
Consultant, Dipont Education Management (China)

(C) 781-775-0433
(F) 617-588-0283
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.debatecoaches.org/pipermail/ndca-l-debatecoaches.org/attachments/20131218/d0483b48/attachment-0004.htm>



More information about the Ndca-l mailing list